fix filenames in split script

2 files changed, 26 insertions, 0 deletions

diff --git a/.gitignore b/.gitignore
index b030227..b2a4598 100644
--- a/.gitignore
+++ b/.gitignore
@@ -12,4 +12,5 @@ data
 output
 build
 dist
+test_data
 
diff --git a/split.sh b/split.sh
new file mode 100755
index 0000000..416d425
--- /dev/null
+++ b/split.sh
@@ -0,0 +1,25 @@
+#/bin/sh
+
+if [ "$#" -ne 3 ]; then
+    echo "Usage: $0 <filename.wav> <chunk size in seconds> <dataset path>"
+    exit
+fi
+
+fn=$1
+chunk_size=$2
+dataset_path=$3
+
+converted=".temp2.wav"
+rm -f $converted
+ffmpeg -y -hide_banner -loglevel error -i $fn -ac 1 -ar 24000 $converted
+
+mkdir $dataset_path
+length=$(ffprobe -i $converted -show_entries format=duration -v quiet -of csv="p=0")
+end=$(echo "$length / $chunk_size - 1" | bc)
+echo "splitting..."
+for i in $(seq 0 $end); do
+    id=$(printf "%04d" $i)
+    ffmpeg -y -hide_banner -loglevel error -ss $(($i * $chunk_size)) -t $chunk_size -i $converted "${dataset_path}/${dataset_path}_${id}.wav"
+done
+echo "done"
+rm -f $converted